Automatic Transcription
Overview
Covers AI-powered speech recognition tools that automatically convert audio and video recordings into text. Key use cases in research include transcription of interviews, focus groups, lectures, and recordings of qualitative fieldwork. Accurate transcription is foundational to qualitative research workflows and reduces a significant time burden on researchers.
Tools in this area
| Tool | Vendor | Maturity | Status | Cost (GBP/yr) |
|---|---|---|---|---|
| openai-whisper | OpenAI | production-ready | adopt | 0 |
| scriber | University of Sheffield (Research Computing) | production-ready | adopt | 0 |
Landscape notes
Sheffield has a solution: scriber is Sheffield’s own on-premise transcription service built on openai-whisper. It is free, available to all Sheffield researchers, and keeps audio data on Sheffield infrastructure. This should be the default recommendation for all Sheffield research transcription needs, particularly where recordings contain personal data or confidential content.
Commercial alternatives ruled out: Commercial transcription services (Sonix, Otter.ai) have previously been ruled out by IT Services due to data governance concerns. scriber was built specifically to address this gap.
Google Meet: Can perform real-time transcription for meetings — acceptable for non-sensitive purposes given Sheffield’s existing Google Workspace license.
Accuracy caveat: All AI transcription requires post-editing for research-quality use. Speaker diarisation (attributing text to specific speakers) remains a limitation of current tools.
Related capability areas
qualitative-coding · data-collection
Sprint history
- 2026-03-04: Initial population from Q1 2026 AI Landscape Map
Navigation
All capability areas: Ideation and Grant Writing · Literature Review · Hypothesis Generation · Data Collection · Automatic Transcription · Autonomous Laboratories · Data Analysis · Scholarly Writing · Qualitative Coding · Knowledge Bases · Research Data Management